Hands-on builds across RAG pipelines, LangChain/LangGraph agents, vLLM deployments, VectorDB search, private LLM stacks (Ollama, LM Studio), LLM fine-tuning, and multimodal AI (TTS, avatars, Stable Diffusion) — with pragmatic notes on orchestration and document automation.
End-to-end RAG with vector search and local inference.
Read MorePEFT/LoRA, GGUF, and private deployment via Ollama.
Read MoreOpenAI-compatible API, long context, low latency.
Read MoreReal-time PPT generation with Model Context Protocol + SSE.
Read More3-agent pipeline (Researcher/Writer/Editor) for SEO posts.
Read MoreStateful, tool-using agent with LangGraph + local LLaMA-3.
Read More